CN107505596B - MIMO active detection signal design and detection system and method based on double extended underwater acoustic channel environment - Google Patents

Abstract

The invention discloses a MIMO active detection signal design and detection system and method based on the double-expanded underwater acoustic channel environment, which is suitable for the field of underwater acoustic detection and positioning under the condition of the fast time-varying double-expanded underwater acoustic channel. The advantage of the ZCZ sequence waveform design based on the periodic autocorrelation and cross-correlation structure proposed by the present invention is that the interference can be realized only by using the periodic cyclic shift correlation characteristic of the waveform sequence without excessive consideration and design of different orthogonal structures. completely eliminated. Through beamforming and matched filtering processing, the virtual data vectors of echo signals of different transmitting array elements are distinguished, the virtual array element aperture is expanded, the target resolution is improved, and the space-time diversity gain of the entire system is improved. All azimuth detection can be achieved at one time, which saves detection time compared with the existing traditional beamforming technology.

Description

Design and detection of MIMO active detection signal based on double extended underwater acoustic channel environment system and method

technical field

The invention relates to a MIMO active detection signal design and detection system and method based on the double-expanded underwater acoustic channel environment, which is suitable for the field of underwater acoustic detection and localization under the condition of the fast time-varying double-expanded underwater acoustic channel.

technical background

The essential difference between the sonar signal and the communication signal is that the communication signal contains all the information of the communication, while the sonar signal has no information and is only a carrier of information. When the transmitted signal of the sonar reaches the detection target, the signal will be reflected. , all the information of the target is contained in the echo signal. The emission waveform of the radar signal is directly related to the information it can carry. The selection of the waveform will directly determine and affect the performance parameters of the sonar system, including signal-to-noise ratio, range resolution, Doppler resolution, time delay and Doppler resolution. Le ambiguity function, etc. The transmission process focuses on how to extract as much useful information of the target as possible from the sonar echo signal. In the waveguide underwater acoustic environment, the channel has fast time-varying characteristics and delay-Doppler double-spreading characteristics, which makes it difficult to extract the target signal. Therefore, it is necessary to reasonably design the sonar signal transmission waveform and signal detection method to overcome the complex environment of the time-varying double extended underwater acoustic channel, improve the space-time diversity gain of the MIMO sonar system, and realize multi-target detection.

SUMMARY OF THE INVENTION

The present invention proposes a MIMO active detection signal design and detection system and method based on a fast time-varying, double extended underwater acoustic channel environment.

Existing MIMO active detection signal designs all adopt aperiodic correlation structures in principle. Although these transmitted signals have good aperiodic autocorrelation and cross-correlation characteristics, they still have certain side lobe interference. The present invention proposes a ZCZ sequence waveform design based on a periodic autocorrelation and cross-correlation structure. The advantage of this waveform is that it does not need to consider and design different orthogonal structures too much, and only uses the periodic cyclic shift correlation of the waveform sequence. This feature can completely eliminate interference. Through beamforming and matched filtering processing, the virtual data vectors of echo signals of different transmitting array elements can be distinguished, the virtual array element aperture can be expanded, the target resolution can be improved, and the space-time diversity gain of the entire system can be improved. All directions can be detected at one time. Saves detection time compared to existing conventional beamforming techniques.

In order to overcome the shortcomings of fast time-varying, large delay and severe Doppler spread of the underwater acoustic channel in the waveguide environment, multi-target detection is realized. The technical scheme adopted by the present invention is: periodic cyclic shift correlation ZCZ sequence design and MIMO underwater acoustic signal detection, and the main devices of the system include:

The ZCZ sequence generator uses cyclic shift and cyclic prefix to generate signals suitable for MIMO active detection;

The transmit power amplifier group, connected with the ZCZ sequence generator, is used for transmit signal power amplification;

The receiving power amplifier group is connected with the receiving hydrophone array to realize the gain control of the receiving signal;

The transmitting transducer array, which is connected with the transmitting power amplifier group, is used to convert the amplified detection signal from an electrical signal to an acoustic signal, and transmit the acoustic signal to the detection water area;

The receiving hydrophone array is used to convert the received echo sound signals into electrical signals;

The coherent processing unit processes the received MIMO signal to obtain the MIMO beam pattern, estimates the MIMO beam pattern, and obtains the potential target azimuth, and then uses the generalized likelihood ratio detector to detect the estimated potential target for the output result of the matched filter. Whether there is a real target in the bearing.

Among them, u _mimo (θ) is the MIMO beam pattern after matching output, θ is the azimuth where the target may exist, H ₁ indicates that there is a real target in the estimated potential target azimuth, and H ₀ indicates that the estimated potential target azimuth does not have a real target. , δ is the detection threshold of the generalized likelihood ratio detector.

Compared with the prior art, the innovation of the present invention is:

The ZCZ transmit signal design based on periodic cyclic shift correlation can resist the waveguide time-varying underwater acoustic channel, and can achieve complete interference cancellation;

Through MIMO array processing, spatial diversity gain can be obtained. Compared with traditional phased array beamforming, the virtual aperture and spatial degrees of freedom are expanded, and narrower main lobes and lower side lobes can be obtained, which improves target detection. resolution and signal-to-noise ratio of echo detection;

Compared with the traditional phased array beam forming, the present invention can realize the detection of all azimuths in one detection, avoids phased scanning, and saves time.

Description of drawings

FIG. 1 is an overall schematic diagram of the underwater acoustic MIMO positioning transceiver system in the present invention.

FIG. 2 is a schematic diagram of a combined transceiver array.

FIG. 3 is a comparison of beam patterns between the MIMO detection system and the phased array detection system in the present invention.

Figure 4 is a schematic diagram of the interference suppression of the ZCZ signal passing through the real waveguide underwater acoustic channel.

Detailed ways

The present invention will be further described below with reference to the accompanying drawings and specific examples, but the implementation and protection scope of the present invention are not limited thereto.

The MIMO underwater acoustic detection device designed based on the ZCZ detection signal of the present invention includes:

The ZCZ sequence generator uses cyclic shift and cyclic prefix to generate signals suitable for MIMO active detection;

The transmit power amplifier group, connected with the ZCZ sequence generator, is used for transmit signal power amplification;

The receiving power amplifier group is connected with the receiving hydrophone array to realize the gain control of the receiving signal;

The transmitting transducer array, which is connected with the transmitting power amplifier group, is used to convert the amplified detection signal from an electrical signal to an acoustic signal, and transmit the acoustic signal to the detection water area;

The receiving hydrophone array is used to convert the received echo sound signals into electrical signals;

The coherent processing unit processes the received MIMO signal to obtain the MIMO beam pattern, estimates the MIMO beam pattern, and obtains the potential target azimuth, and then uses the generalized likelihood ratio detector to detect the estimated potential target for the output result of the matched filter. Whether there is a real target in the bearing.

1 shows the working principle diagram of the signal generation and detection method of the present invention, including: ZCZ sequence generator, transmit power amplifier group, receive power amplifier group, transmit transducer array, receive hydrophone array, and coherent processing unit.

The layout of the transmitting array and the receiving array is shown in Figure 2. The number of transmitting array elements is N _t =6, the number of receiving array elements is N _r =7, the distance between transmitting array elements is 10cm, and the distance between receiving array elements is 7cm.

It should be noted that although Fig. 1 uses N _t =6 and N _r =7 for more concise expression, in the present invention, the transmitting transducer array N _t and the receiving transducer array N _r are not limited to this special case , as long as the transmit and receive conditions allow, a reasonable beam pattern can be obtained by adjusting the number of transmit and receive array elements.

Generating a sounding signal by using the MIMO sounding signal generating device of the present invention includes the following steps:

Step 1: For a ZCZ sequence structure, ^Fn represents a cluster with M ZCZ sequences, each ZCZ sequence has a length of L, which can be further expressed as F(L, M, Z _CZ ), Z _CZ zero correlation interval length , when n is 0, the fundamental matrix used to generate the ZCZ sequence can be expressed as:

Step 2: Further generalization, L ₀ =2 is the starting sequence length, and let the starting n=1, the longer ZCZ sequence matrix can be expressed as: F(L, M, Z _CZ )=(2 ²ⁿ L ₀ , ²ⁿ⁺ 1,2n ⁺ 1)=(8,4,3). The structure can be represented by formula (4):

表示对

取反，F¹的每一行都是一组ZCZ序列。The matrix in formula (4)

express right

Negated, each row of F1 is ^a set of ZCZ sequences.

Step 3: Generalize the ZCZ sequence cluster, for F ^n-1 = (2 ^2(n-1) L ₀ , 2 ⁿ , 2 ^n-1 +1), a larger sequence cluster F ⁿ = (2 ²ⁿ L ₀ ,2 ⁿ⁺¹ ,2 ⁿ +1) can be expressed as:

和

可以用式(6)处理方式来实现：In formula (5),

and

It can be realized by the processing method of formula (6):

The ZCZ signal generated by formula (5) is processed as follows:

Step 1: Perform cyclic shift processing on the generated one-way ZCZ sequence, and the number of signals generated by the shift is determined by the number of transmitting transducer array elements.

Step 2: In order to ensure that the generated signal sequence has a good cyclic shift autocorrelation characteristic, it is necessary to add a cyclic prefix to the generated signal of each channel.

The ZCZ signal processed by cyclic shift and cyclic prefix is integrated by the transmitting driving vector, and the signal is amplified by the transmitting power amplifier group. Finally, the MIMO transducer converts the electrical signal into an acoustic signal and transmits it.

The entire signal transmission and detection process includes the following steps:

Step 1: The signal transmitter transmits a set of M ZCZ cyclic shift sequences processed by cyclic shift + cyclic prefix, and the signal is expressed in vector form as s[n]=(s ₁ [n], s ₂ [n], …,s _M [n]) ^T , the covariance matrix of the transmitted signal can be expressed as:

In formula (7), M≥1 is a positive integer, n represents the sampling point at the nth time of the quadrature signal, 1≤n≤L, L represents the length of the codeword sequence obtained by sampling, and _βij represents the i-th transmission signal s The cross-correlation coefficient between _i [n] and the jth transmitted signal s _j [n], s ^H [n] is the s[n] conjugate transpose operation, R _s can be decomposed by SVD as:

R _s = UΛU ^H (8)

where U is a unitary matrix, Λ is a diagonal matrix, and U ^H is the conjugate transpose of U. When the transmitted signals are completely orthogonal, R _s = _IM .

Step 2: The detection system is composed of N _t transmitting array elements and N _r receiving array elements. The transmitting array steering vector a _t (θ) and the receiving array steering vector a _r (θ) are expressed as Eq. (9) and Eq. (10):

where at (θ) is the steering vector of the transmitting transducer array, a _r (θ ₎ is the steering vector of the receiving transducer array, θ is the potential target azimuth angle, and N _t is the number of elements of the transmitting transducer array N _r is the number of array elements of the receiving hydrophone array, d _t is the distance between the transmitting transducer array elements, d _r is the receiving hydrophone array element distance, f is the center frequency of the transmitting signal of the transmitter, and c is the water Speed of sound, ( ) ^T represents the matrix transpose operation.

Step 3: The N _t quadrature signals s[n] formed by the driving vector beam are sent to the transmitting transducer array through the power amplifier group, and the transmitting transducer array converts the amplified quadrature signals from electrical signals into acoustic signals. The signal is sent to the water area to be detected; the signal is reflected by the target to be detected, the echo signal is received by the receiving hydrophone array, and the echo acoustic signal is converted into an electrical signal by the receiving hydrophone power amplifier group. The received signal can be expressed as:

的复高斯分布，其中

是噪声功率，

是秩为N_r的单位矩阵。In formula (11), r[n] is the electrical signal received by the receiving hydrophone, r[n]=[r ₁ [n],r ₂ [n],...,r _Nr [n]] ^T , n =1,2,...L is the signal sequence received by the N _r -element receiving array, α(θ) is the signal propagation attenuation coefficient, w[n] is the signal group received by the receiving hydrophone array is not related to the transmitting signal group The additive noise vector of . obey

The complex Gaussian distribution of , where

is the noise power,

is an identity matrix of rank _Nr .

The received signal sequence of length L is expressed in the form of a matrix, which takes the form of array data storage:

N＝[w[1],w[2],…，w[L]]，

where R=[r[1],r[2],…,r[L]], S=[s[1],s[2],…s[L]], then

N=[w[1],w[2],...,w[L]],

做匹配滤波处理来获得，通过匹配滤波处理得到的充分检验统计矩阵为：Step 4: Detect and locate the target according to the MIMO received signal model r[n] described by Equation (11).

It is obtained by doing matched filtering processing, and the sufficient test statistical matrix obtained by matching filtering processing is:

The transmitted signal has perfect cyclic shift autocorrelation characteristics, and the full test statistic matrix Y _mimo is further simplified as:

Perform column quantization processing on the sufficient statistic obtained by Equation (14), and obtain the column quantized representation of the sufficient statistic shown in Equation (15):

y _mimo =vec(Y _mimo )=α(θ ₀ )d(θ ₀ )+v (15)

是长度为N_tN_r×1的接收水听器阵列匹配输出响应，

是克罗内克积，

是服从

的复高斯噪声，其中

是向量化后的噪声功率，

是秩为MN的单位矩阵。In formula (15), θ ₀ is the potential azimuth of the target, α(θ ₀ ) is the signal propagation attenuation coefficient of the potential target azimuth angle,

is the matched output response of the receiving hydrophone array of length N _t N _r × 1,

is the Kronecker product,

is to obey

complex Gaussian noise, where

is the quantized noise power,

is the identity matrix of rank MN.

Step 5: According to the sufficient statistic y _mimo obtained from Equation (15), the system transmit and receive beam pattern and the maximum likelihood estimate containing the potential target azimuth are obtained from Equation (16).

表示对a_r(θ)做共轭转置运算，

表示对a_t(θ)做共轭转置运算，

表示潜在目标所在方位的最大似然估计。In formula (16), u _mimo (θ) is the MIMO mode after the matching output, θ is the orientation of the possible target, || represents the absolute value operation, |||| represents the vector modulo operation,

Indicates that the conjugate transpose operation is performed on a _r (θ),

represents the conjugate transpose operation on a _t (θ),

Represents the maximum likelihood estimate of where the potential target is located.

Using the beam pattern expressed by Equation (16), for each -90°≤θ≤90°, calculate the corresponding beam pattern value, and obtain the beam pattern diagram shown by the solid line in Figure 3. According to Equation (17) and Beam pattern map, search for the angle θ corresponding to the maximum value of u _mim o(θ), which is the estimated azimuth of the potential target

上是否真实存在目标：Step 6: Use equation (17) to estimate the MIMO beam pattern to obtain the potential target azimuth, and then use the generalized likelihood ratio detector of equation (18) to detect the potential target azimuth

Is there a real target on:

是估计出的潜在目标方位上的波束模式，H₁表示估计出的潜在目标方位存在真实的目标，H₀表示估计出的潜在目标方位不存在真实的目标，δ为广义似然比检测器的检测阈值。由预先设定的虚警概率P_f来决定，当

时认为潜在目标方位

目标真实存在，否则认为潜在目标方位

目标并不真实存在。In formula (18),

is the estimated beam pattern on the potential target azimuth, H ₁ indicates that there is a real target in the estimated potential target azimuth, H ₀ indicates that there is no real target in the estimated potential target azimuth, and δ is the generalized likelihood ratio detector. detection threshold. is determined by the preset false alarm probability P _f , when

potential target orientation

The target is real, otherwise it is considered the potential target orientation

The goal doesn't really exist.

与实际情况中待检目标方位相符。As can be seen from Figure 3, the estimated target orientation

It is consistent with the orientation of the target to be inspected in the actual situation.

Compared with the beam-forming beam pattern of the conventional phased array, the beam pattern of the detection method of the present invention has narrower main lobes and narrower side lobes, and the first side lobes are 13 dB lower than the first side lobes of the conventional phased array.

The narrower main lobe of the beam mode can achieve higher target resolution, and the lower side lobes can reduce the interference of background noise and false alarms, improve the signal-to-noise ratio of the received signal, and improve the target detection probability.

The signal detection based on the cyclic shift ZCZ sequence proposed by the present invention can resist the waveguide time-varying double extended underwater acoustic channel, and can realize the complete elimination of interference. Figure 4 shows that the cyclic shift related ZCZ signal passes through the real underwater acoustic channel. Compared with the traditional pseudo-random sequence, the cyclic shift-correlated ZCZ sequence can completely eliminate the interference and has better detection performance.

Claims (4)

1. A design and detection method of a MIMO active detection signal design and detection system based on a double-extension underwater acoustic channel environment is provided, wherein the design and detection system comprises:

the ZCZ sequence generator generates a signal suitable for MIMO active detection by adopting a cyclic shift and cyclic prefix mode;

the transmitting power amplifier group is connected with the ZCZ sequence generator and used for transmitting signal power amplification;

the transmitting transducer array is connected with the transmitting power amplifier group and used for converting the amplified detection signal into an acoustic signal from an electric signal and transmitting the acoustic signal to a detection water area;

the receiving hydrophone array is used for converting the received echo acoustic signals into electric signals;

the receiving power amplifier group is connected with the receiving hydrophone array and used for realizing the gain control of the received signals;

the coherent processing unit is used for processing the received MIMO signals to obtain an MIMO beam mode, estimating the MIMO beam mode to obtain a potential target azimuth, and then detecting whether a real target exists in the estimated potential target azimuth or not by utilizing a generalized likelihood ratio detector according to a result output by matched filtering;

the method is characterized by comprising the following steps:

1) a ZCZ sequence generator generates a ZCZ sequence; performing cyclic shift processing on the generated ZCZ sequence, determining several paths of signals generated by shift according to the number of elements of a transmitting transducer, and adding cyclic prefix processing to each path of generated signals;

2) the ZCZ signals subjected to cyclic shift and cyclic prefix processing are subjected to transmission driving vector integration, the signals are amplified by a transmission power amplifier group, the amplified signals are sent to a transmission transducer array by the transmission power amplifier group, and the electric signals are converted into acoustic signals by the transmission transducer array and sent to a water area to be detected;

3) the signal is reflected by a target to be detected, the echo signal is received by the receiving hydrophone array, the echo acoustic signal is converted into an electric signal by the receiving power amplifier group, the coherent processing unit estimates the MIMO wave beam mode to obtain a potential target azimuth, and then the generalized likelihood ratio detector is utilized to detect whether the target really exists in the potential target azimuth.

2. The method for designing and detecting a MIMO active probing signal based on a dual-extension underwater acoustic channel environment according to claim 1, wherein the step 1) specifically comprises:

step 1.1: for a ZCZ sequence structure, FⁿRepresents a cluster of M ZCZ sequences, each ZCZ sequence having a length of L, FⁿFurther denoted as F (L, M, Z)_CZ)，Z_CZFor a zero correlation interval length, when n is 0, the base matrix for generating ZCZ sequences is represented as:

F(L,M,Z_CZ)＝(2,2,1)(1)

step 1.2: with L₀The ZCZ sequence matrix, with starting sequence length of 2 and starting n of 1, which is longer, is represented as:

F(L,M,Z_CZ)＝(2²ⁿL₀,2ⁿ⁺¹,2ⁿ+1) — (8,4,3), the structure is represented by equation (2):

in the formula (2), matrix

Presentation pair

Taking the inverse, F¹Each row of (a) is a set of ZCZ sequences;

step 1.3: generalizing ZCZ sequence clusters for F^n-1＝(2^2(n-1)L₀,2ⁿ,2^n-1+1),Fⁿ＝(2²ⁿL₀,2ⁿ⁺¹,2ⁿ+1), a larger cluster Fⁿ＝(2²ⁿL₀,2ⁿ⁺¹,2ⁿ+1) is expressed as:

step 1.4: in the formula (3), the reaction mixture is,

and

is realized by the processing mode of the formula (4)

3. The method for designing and detecting a MIMO active probing signal based on a dual-extension underwater acoustic channel environment according to claim 1, wherein the step 2) specifically comprises:

step 2.1: the signal transmitter transmits a group of M ZCZ cyclic shift sequences processed by cyclic shift and cyclic prefix, and the signal is expressed as s [ n ] in a vector form]＝(s₁[n],s₂[n],…,s_M[n])^TThe covariance matrix of the transmitted signal is expressed as:

in the formula (5), M is a positive integer larger than or equal to 1, n represents a sampling point at the nth moment of the orthogonal signal, n is larger than or equal to 1 and smaller than or equal to L, β_ijRepresenting the ith transmission signal s_i[n]And j-th transmission signal s_j[n]Cross correlation coefficient between, s^H[n]Is s [ n ]]Conjugate transpose operation, R_sDecomposition by SVD is:

R_s＝UΛU^H(6)

in the formula (6), U is unitary matrix, Λ is diagonal matrix, and U is^HFor the conjugate transpose of U, R is when the transmitted signals are perfectly orthogonal_s＝I_M；

Step 2.2: transmit array steering vector a_t(theta) and receiving array steering vector a_r(theta) is represented by (7)

And (8):

wherein a is_t(theta) is the steering vector of the transmit transducer array, a_r(theta) is the steering vector of the receive transducer array, theta is the potential target azimuth angle, N_tIs the number of array elements of the transmitting transducer array, N_rFor receiving the number of array elements of the hydrophone array, d_tTo transmit transducer element spacing, d_rFor receiving hydrophone array element spacing, f is transmitter transmit signal center frequency, c is underwater sound velocity, (.)^TRepresenting a matrix transposition operation;

step 2.3: n via steering vector beamforming_tA quadrature signal s [ n ]]The orthogonal signals are transmitted to a transmitting transducer array through a power amplifier group, the transmitting transducer array converts the amplified orthogonal signals into acoustic signals from electric signals, and the acoustic signals are transmitted to a water area to be detected; the signal is reflected by a target to be detected, an echo signal is received by the receiving hydrophone array, and the echo sound signal is converted into an electric signal by the receiving hydrophone power amplifier group.

4. The method for designing and detecting a MIMO active probing signal based on a dual-extension underwater acoustic channel environment according to claim 1, wherein the step 3) specifically comprises:

step 3.1: the signals received by the receiving hydrophones are expressed as:

in formula (9), r [ n ]]In order to receive the electrical signals received by the hydrophones,

n is 1,2, … L is N_rThe signal sequence received by the element receiving array, α (theta) is the signal propagation attenuation coefficient, w [ n ]]Additive noise direction for receiving signals received by a hydrophone array and uncorrelated with the transmitted signalsAmount, compliance

A complex Gaussian distribution of wherein

Is the power of the noise or noise,

is rank of N_rThe identity matrix of (1);

step 3.2: MIMO received signal model r [ n ] described by equation (9)]To detect and locate the target by cyclically shifting the signal with ZCZ

Performing matched filtering to obtain a sufficient test statistical matrix, wherein the sufficient test statistical matrix obtained through the matched filtering is as follows:

the transmitted signal has perfect circular shift autocorrelation characteristic and fully checks the statistical matrix Y_mimoFurther simplification is as follows:

performing column vectorization processing on the sufficient statistics obtained by the formula (11) to obtain column vectorization expression of the sufficient statistics shown by the formula (12):

y_mimo＝vec(Y_mimo)＝α(θ₀)d(θ₀)+v (12)

in the formula (12), θ₀Is the target potential bearing, α (θ)₀) Signal propagation attenuation coefficients for the potential target azimuth angles,

is of length N_tN_r× 1 match the output response,

is the product of the kronecker product,

is subject to

Of complex Gaussian, wherein

Is the power of the noise or noise,

is rank of N_tN_rThe identity matrix of (1);

step 3.3: the sufficient statistical vector y obtained according to step 3.2_mimoObtaining the system transmit-receive beam pattern and the maximum likelihood estimation containing the potential target position by the formula (13)

In the formula (13), u_mimo(θ) is the MIMO beam pattern after matching output, | | | | represents absolute value computation, | | | | represents vector modulo computation,

represents a pair of_r(theta) performing a conjugate transpose operation,

represents a pair of_t(theta) performing a conjugate transpose operation,

a maximum likelihood estimate representing the orientation of the potential target;

step 3.4: estimating the MIMO beam mode by using the formula (13) to obtain a potential target azimuth, and then detecting whether the target really exists in the potential target azimuth by using the generalized likelihood ratio detector of the formula (14):

in the formula (14), the compound represented by the formula (I),

for the estimated beam pattern at the location of the potential target, H₁Indicating that the estimated orientation of the potential target is a true target, H₀And the estimated potential target azimuth is not provided with a real target and is the detection threshold of the generalized likelihood ratio detector.

Images